Angel Alique

Fuzzy model and hierarchical fuzzy control integration: an approach for milling process optimization

Abstract Process optimization is a very important subject to several industrial sectors in confronting the growth on markets competition. However, due to the complexity of some processes, their optimization is not an easy task; therefore, to accomplish this objective, intelligent techniques should be used. We are working on end-milling process optimization through combining analytic and fuzzy techniques. This paper describes in a general form a hierarchical structure of fuzzy control and fuzzy model used in end-milling process. These modules represent possible solutions to complex problems: optimization and supervision of milling process. The articles goal is to discuss some results of integrated modules in on-line operation.

Toward intelligent machining: hierarchical fuzzy control for the end milling process

The difficulties in implementing adaptive and other advanced control schemes in industrial machining processes have encouraged researchers to combine the utilization of one hierarchical level, a fuzzy control algorithm, and robust sensing systems. The main idea of this paper deals with self-regulating controllers (SRCs). The control signals scaling factor (output scaling factor) is self-regulated during the control process, and it can assure the optimum gain setting for the hierarchical fuzzy controller. An important role in this strategy is performed by a robust sensing system based on current sensors. For comparison, the CNC-PLCs own control loops, a hierarchical fuzzy controller based on look-up tables, and the hierarchical fuzzy controller with a self-regulating output scaling factor GC are studied. The performances of these controllers are compared. The results indicate that the hierarchical fuzzy controller with a self-regulating output scaling factor yields the best performances among them. The index known as the metal removal rate is increased, and the in-process time is reduced by 50%. Thus, higher production rates are obtained. The hierarchical fuzzy controller is equipped with three basic requirements: flexibility, low cost, and compatibility with any CNC manufacturer.

Embedded fuzzy-control system for machining processes results of a case study

In this paper a fuzzy-control system has been designed, implemented and embedded in an open CNC. The integration process, design steps and results of applying an embedded fuzzy-control system are shown through the example of real machining operations. The controller uses internal CNC signals (i.e. spindle-motor current) that are gathered and mathematically processed by means of an integrated application. The results show that, at least in rough milling operations, internal CNC signals can double as an intelligent, sensorless control system. Actual industrial tests show a higher machining efficiency (i.e. in-process time is reduced by 10% and total estimated savings the system would provide are about 78%).

CORBA-Based Open Platform for Processes Monitoring. An Application to a Complex Electromechanical Process

The goal of this work is to develop an open software platform called SYNERGY, supported by portable, low cost and worldwide-accepted technologies (i.e., Real Time CORBA), focused on networked control systems. Preliminary results of SYNERGY corroborate the viability for networked control, supervision and monitoring of complex electromechanical processes like high speed machining (HSM), on the basis of current communications and computation technologies upon open architectures.

Controlling a complex electromechanical process on the basis of a neurofuzzy approach

This paper shows the viability of implementing a control strategy based on the internal-model control paradigm, which is a useful synergy of a dynamic ANN trained from real-life data and used to predict process output and a fuzzy-logic control (FLC) that enhances the control systems overall performance. A force control problem involving a complex electromechanical system, represented here by the machining process, is considered as a case study. The main goal is to control a single-output variable, cutting force, by changing a single-input variable, feed rate. The proposed neurofuzzy-control (NFC) scheme consists of a dynamic model using ANNs to estimate process output, and a fuzzy-logic controller (FLC) with the same static gain as the inverse model to determine the control inputs (feed rate) necessary to keep the cutting force constant. Four approaches, the fuzzy-logic controller (FLC), the direct inverse controller based on ANNs (DIC-NN), the internal-model controller (IMC-NN) and a neurofuzzy controller (NFC), are simulated and their performances are assessed in terms of several performance measurements. The results demonstrate that the NFC strategy provides better disturbance rejection than the IMC-NN and the FLC for the cases analyzed.

Dynamic Model of the Machining Process on the Basis of Neural Networks: From Simulation to Real Time Application

Nowadays, the modeling of complex manufacturing tasks is a key issue. In this work, as a case study is selected the application of a dynamic model to predict cutting force in machining processes. A model created using Artificial Neural Networks (ANN), able to predict the process output is introduced in order to deal with the characteristics of such an ill-defined process. This model describes the dynamic response of the output before changes in the process input command (feed rate) and process parameters (depth of cut). Experimental tests are made in a professional machining centre, with different cutting conditions, on real time data. The model provides sufficiently accurate prediction of cutting force, since the process-dependent specific dynamic properties are adequately reflected.

Hierarchical fuzzy control of the milling process with a self-tuning algorithm

This paper presents a hierarchical fuzzy logic controller (HFLC) with a self-tuning (ST) procedure based on pattern recognition of the closed-loop system response. A real-time tuning of the controller scaling factors is performed on the basis of the measured peaks in the error signal. In order to demonstrate the improved performance and effectiveness of this scheme, the ST-HFLC is applied to the end milling process. The response of the controlled process using the proposed controller (ST-HFLC), a standard HFLC and an HFLC with a self-regulating output scaling factor is analyzed. The ST-HFLC provides a better transient performance than the others. Without significant variation in rise time, a non-oscillating system with a short settling time can be achieved. These positive features, in the case of our particular application, may represent an increase of cutting tool life as well as the avoidance of chatter marks due to inappropriate cutting conditions.

Multivariable circle criterion: stable fuzzy control of a milling process

Finding for a complex control plant an analytical model that is detailed enough for controller design is often hard or even impossible. In such cases the regulation of the system can be performed by knowledge-based fuzzy control. Under the assumption that there exists a locally valid linear description for the plant the stability problem of the nonlinear fuzzy control-loop can be addressed with the multivariable circle criterion, e.g. in linear-matrix-inequality format. The application of that stability criterion is demonstrated for fuzzy control of a nonlinear parameter-varying milling process and approved in both simulations and real-time application.

A neural network-based model for the prediction of cutting force in milling process. A progress study on a real case

In spite of recent developments focusing on milling process optimization through an effective cutting force control, there is a need for the analysis of the transient response of these systems because undesirable oscillations in cutting force can be harmful to the quality of the finishing surface and tools. The main goal of this work is to develop a versatile neural network model which can online predict the mean cutting force under commonly encountered conditions. Using this model, easily obtained from a straightforward machining test, developments of complex adaptive controllers and monitoring systems can be carried out. As a result, a good model for predicting the cutting process was obtained.

Embedded fuzzy control system: application to an electromechanical system

Nowadays with open computerized numerical controls internal control signals can be gathered and mathematically processed by means of integrated applications. Working with a commercial open computerized numerical control, a fuzzy control system has been designed, implemented and embedded that can provide an additional optimization function for cutting speed. The results show that, at least in rough milling operations, internal signals can double as an intelligent, sensorless control system. The integration process, design steps and results of applying an embedded fuzzy control system are shown through the example of real machining operations.